Synthesis, structural and spectroscopic studies on the lanthanoid p-aminobenzoates and derived optically functional polyurethane composites

Article abstract of DOI:10.1002/ejic.200600797

Aromatic carboxylates of the rare earths are known for highly efficient luminescence, thus holding the promise of applicability in optical devices, but may be problematic to apply as such due to their structure, which is polymeric in nature. Therefore, means of conversion into molecular species or eventually polymer embeddings are sought, while at the same time maintaining their advantageous optical features. For these purposes, co-ligations of rare earth p-aminobenzoates with bidentate bipyridine and phenanthroline, tridentate terpyridine and polyurethanes, which may also be perceived as polydentate ligands, have been carried out. The materials were characterised with focus on their structure and optical performance. Brightly luminescing composites were obtained from reactions with aromatic and aliphatic diisocyanates. Prolonged UV-irradiation of the materials proved hexamethylene diisocyanates to be superior with regard to photodegradation. Transparent polyurethane monoliths with emission quantum yields between 40 % and 55 % were eventually prepared from covalent attachement of Tb(p-aminobenzoate)₃ and its bipyridine co-ligated complexes to a hexamethylene diisocyanate/polyol system. Wiley-VCH Verlag GmbH & Co. KGaA, 2007.

Full text of DOI:10.1002/ejic.200600797

FULL PAPER
DOI: 10.1002/ejic.200600797
Synthesis, Structural and Spectroscopic Studies on the Lanthanoid
p-Aminobenzoates and Derived Optically Functional Polyurethane Composites
Timothy Fiedler,[a][‡] Matthias Hilder,[b] Peter C. Junk,*[b] Ulrich H. Kynast,[a]
Marina M. Lezhnina,*[a][‡‡] and Marek Warzala[a]
Keywords: Luminescence / Lanthanoids / Polyurethane composites / Carboxylates
Aromatic carboxylates of the rare earths are known for
highly efficient luminescence, thus holding the promise of
applicability in optical devices, but may be problematic to
apply as such due to their structure, which is polymeric in
nature. Therefore, means of conversion into molecular spe-
cies or eventually polymer embeddings are sought, while at
the same time maintaining their advantageous optical fea-
tures. For these purposes, co-ligations of rare earth p-amino-
benzoates with bidentate bipyridine and phenanthroline, tri-
dentate terpyridine and polyurethanes, which may also be
perceived as polydentate ligands, have been carried out. The
materials were characterised with focus on their structure
and optical performance. Brightly luminescing composites
were obtained from reactions with aromatic and aliphatic di-
isocyanates. Prolonged UV-irradiation of the materials
proved hexamethylene diisocyanates to be superior with re-
gard to photodegradation. Transparent polyurethane mono-
liths with emission quantum yields between 40% and 55%
were eventually prepared from covalent attachement of
Tb(p-aminobenzoate)3 and its bipyridine co-ligated com-
plexes to a hexamethylene diisocyanate/polyol system.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2007)
This lack of understanding is particularly true for the
group of aromatic carboxylates imbedded in mentioned
hosting matrices, which includes some of the most ef-
ficiently luminescing rare earth species. Rare earth aromatic
carboxylates generally form coordination-polymeric associ-
ates with corresponding short-range interactions rather
than isolated molecular entities.[6] Resulting insolubility and
low vapour pressure pose a practical shortcoming with re-
gard to matrix incorporation schemes, as the destruction of
the original rare earth coordination polymer often quenches
or diminishes the luminescence efficiencies. Bipyridines
(“bipy”) and phenanthrolines (“phen”), acting both as co-
chelating and co-sensitizing ligands, can help to circumvent
the solubility or volatility problem by forming new molecu-
lar coordination complexes. At the same time, they can help
to keep the immediate chemical environment of the ion free
from water, which is frequently encountered in the pure car-
boxylates. In addition to these bidentate co-ligands, we have
additionally used terpyridine (“terpy”) as a co-ligand. Also,
some matrices may analogously be understood in terms of
co-ligation, in that they may also occupy the rare earth co-
ordination sphere in polydentate fashion as well, e.g. oxygen
atoms of silica based matrices or silicones.
1. Introduction
Efficiently luminescing complexes of the rare earths are
not only spectacular visually, but also of interest in numer-
ous photophysical applications, ranging from sensors over
light emitting devices (LEDs, OLEDs) to luminescent
immunoassays. Practically, it is frequently desirable to in-
corporate them into matrices in order to render them appli-
cable, or even appropriately functional in corresponding de-
vices. As such, sol-gel matrices, zeolites and various poly-
mers have been utilised.
Although the principle “local” intramolecular lumines-
cence mechanism of the complexes is understood compara-
bly well, especially in β-diketonates,[1–5] a reliable and com-
prehensive description of the structural factors influencing
the optical performance appears to be lacking, especially
with regard to carboxylates.
[a] University of Applied Sciences Muenster, Dep. of Chemical
Engineering/Applied Materials Sciences,
Stegerwaldstr. 39, 48565 Steinfurt, Germany
Fax: +49-2551-962-187
E-mail: marina@fh-muenster.de
[b] School of Chemistry, Box 23, Monash University,
Clayton, Vic. 3800, Australia
Fax: +61-03-9905-4597
E-mail: peter.junk@sci.monash.edu.au
[‡] Current address: OSRAM GmbH, Phosphor Development &
Applications,
In this report, we focus on p-aminobenzoic acid (“pa-
baH”), because we learnt from single crystal structure data
that the amino group might be chemically accessible for co-
valent attachment to a matrix on one hand, and, e.g. Tb(p-
aminobenzoate)3 [“Tb(paba)3”], show intense luminescence
Hellabrunner Straße 1, 81543 München
[‡‡]On leave from: Mari Technical State University Yoshkar-Ola,
Institute of Physics,
Lenin-pl. 3, 424000 Yoshkar-Ola, Russia
Eur. J. Inorg. Chem. 2007, 291–301
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
291

P. C. Junk, M. M. Lezhnina et al.
FULL PAPER
as the pure coordination polymer on the other hand, which
can be enhanced by co-coordination,[7] and even be main-
tained on incorporation into zeolite matrix.[8] While this re-
search was in progress, some of the crystal structures were
confirmed by reports from other groups, or were previously
published.[9] We will additionally provide results on the in-
teraction of Tb(paba)3, Tb(paba)3phen and Tb(paba)3bipy
with isocyanates and eventually a polyurethane matrix and
their behaviour under prolonged UV irradiation.
(1)
(2)
The FTIR spectra of all compounds showed significant
changes in the carbonyl stretching frequency compared
with the free ligand (see the representative representation of
[Eu(paba)3H2O], and [Tb(paba)3H2O] in comparison with
H(paba) in Figure 1). In p-aminobenzoic acid, the C=O
stretch occurs at 1655 cm–1, whereas in the lanthanoid com-
plexes, the band is shifted to significantly lower wave num-
2. Results and Discussion
2.1. Coordination Polymers
A [Ln(paba)3·H2O]n series with the members [Ln = Eu bers and is split into two distinct bands around 1390 and
(1), Gd (2) and Tb (3); pabaH = p-aminobenzoic acid], were 1500 cm–1. In the compounds 1–3 there are increased ab-
conveniently synthesised in good yield by treatment of an sorbances in the 3300–3450 cm–1 region attributable to
aqueous solution of the respective lanthanoid chloride with water in the metal complexes. The complexes 1–3 were pure
sodium p-aminobenzoate [Equation (1)]. We could also ob- by Na2EDTA titration for lanthanoid content and carbon
tain a modified phase of the terbium complex, being the microanalysis for the composition [Ln(paba)3H2O] as sug-
dimeric [Tb(paba)3(H2O)2]2·2H2O (4), which was obtained gested by the X-ray crystal structures (below).
in a similar manner [Equation (2)], but in the presence of
To ascertain the thermal stability of a representative sam-
1,10-phenanthroline, in an attempt to prepare phenan- ple of the [Ln(paba)3H2O] series, the Eu analogue was ana-
throline co-coordinated complexes of the series. In the case lysed by thermogravimetric analysis. The exothermic release
of compound 4, phenanthroline was not incorporated in of water (peak maximum at 200 °C), which takes place in
the final product, rather a different structural phase (of a the temperature range of 120–220 °C is accompanied by a
previously published structure[9d]) of the terbium p-amino- weight loss of 3% which is identical to the weight loss ex-
benzoate was isolated. This was therefore not characterised pected for the stated composition. This reaction can be de-
further because it was a previously known compound.
scribed by Equation (3).
Figure 1. FTIR spectra of H(paba) (bottom), Tb(paba)3H2O (middle) and Eu(paba)3H2O (top).
292
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 291–301

Optically Functional Polyurethane Composites
FULL PAPER
Eu(C6H4NH2COO)3(H2O)ǞEu(C6H4NH2COO)3 +H2O
(3)
for this series of compounds are approximately 4.50 to
4.80 Å and for compound 2, this distance is 4.725(3) Å.
The thermal decay starts at 330 °C and continues to
810 °C. It is accompanied by several endothermic events
(390, 430, 500, and 740 °C). Although the slope of the TG
curve is very steady, there is a slight change in the gradient
(330–460 °C) accompanied by a weight loss of 30%. This is
the weight loss expected for the formation of the dimeric
europium benzoate complex connected by an oxo bridge. A
plausible reaction sequence is shown in Equations (4), (5)
and (6).
Eu(C6H4NH2COO)3 +3 O2 ǞEu(C6H5COO)3 +1.5 H2O+3 NO
(4)
Eu(C6H5COO)3 +7.5 O2 ǞEu(C6H5COO)2(OH)+2 H2O+7 CO2
(5)
2 Eu(C6H5COO)2(OH)Ǟ
(C6H5COO)2Eu–O–Eu(C6H5COO)2 +H2O (6)
In the range of 460–810 °C the final oxidation product is
formed. The total mass loss of 70% is very close to the
expected (69%) as shown in Equation (7).
(C6H5COO)2Eu–O–Eu(C6H5COO)2 +30.5 O2 Ǟ
Eu2O3 +10 H2O+28 CO2 (7)
Single crystals of compound 2 were grown by slow con-
centration of aqueous samples and its solid-state structures
determined using X-ray crystallography. [Gd(paba)3H2O]n
is isomorphous and structurally identical to that previously
published for [Ln(paba)3H2O]n Ln = La, Ce, Pr, Nd, Sm,
Eu, Tb, Dy, Er,[9] effectively establishing this structural type
for the lighter and larger lanthanoids extends to Ln = Er.
Compound 2 crystallises in the monoclinic space group
P21/n with one whole monomeric unit in the asymmetric
unit. The lanthanoid metal centre is eight coordinate being
Figure 2. a) X-ray crystal structure of coordination sphere about
Gd in polymeric [Gd(paba)3]·H2O (2). Thermal ellipsoids are at
the 50% probability level. Selected bond lengths [Å]: Gd(1)–O(1)
2.461(2), Gd(1)–O(2) 2.455(2), Gd(1)–O(3) 2.336(2), Gd(1)–O(4)
2.304(2), Gd(1)–O(5) 2.342(2), Gd(1)–O(6) 2.349(3), Gd(1)–O(7)
2.513(3), Gd(1)–N(3) 2.668(3). b) X-ray crystal structure of poly-
meric [Gd(paba)3]·H2O (2). Thermal ellipsoids are at the 50%
probability level.
Single crystals of compound 6 suitable for X-ray diffrac-
bound by one chelating carboxylate, four monodentate car- tion studies were grown from an aqueous solution of
boxylates, one water molecule and perhaps surprisingly, one [Tb(paba)3H2O]n in the presence of 1,10-phenanthroline.
N(amino) group [Figure 2, a)]. Each of the monodentate car- The structure of this phase of the terbium analogue is iso-
boxylate groups bridge to an adjacent Ln centre, while the morphous to that established for [Ln(paba)3(H2O)2]2·2H2O
chelating carboxylate is solely bound to one Ln centre. The (Ln = Tb,[9d] Dy,[11] Ho,[9d] Y, [9d] Yb,[9d,12] Lu[9d]) and thus
overall structure is thus a three-dimensional polymeric spe- an example of the heavier structural type for the p-amino-
cies [Figure 2, b)] which is additionally extended through benzoates of the lanthanoids and will not be discussed fur-
O–H···O and N–H···O hydrogen bonds. The Ln–O and Ln– ther. Significantly, by crystallisation of the terbium ana-
N bond lengths (see Figure 2, caption) demonstrates well logue as both structural phases, the structural boundary for
the lanthanoid contraction when compared with the other the lanthanoid p-aminobenzoates existing as a dimer has
lanthanoid analogues,[9] with all internuclear distances now been established to start at Tb ranging to Lu, while
about the Ln centres diminishing across the series in line the polymeric phase extends from La to Er. The reason for
with the reduction in ionic radii of the metal.[10] It is rather the isolation of two different phases presumably lies in the
surprising that the lanthanoid centre is coordinated by an mode of crystallisation. We isolated the dimer phase in the
amine, given the compound was crystallised from water. It presence of 1,10-phenanthroline and this presumably affects
may be expected that water would complete the coordina- the solubility of the polymeric phase.
tion sphere of the oxophilic rare-earth metal and presum-
X-ray powder diffraction patterns of bulk material of
ably, binding of the amine to the metal, arises from a pack- compounds 1 to 3 were compared with patterns calculated
ing effect. It is worthy to mention the Ln···Ln internuclear from X-ray single crystal experiments and established that
distances, particularly with respect to the possibility of the structure of the bulk materials in all cases was identical
Ln···Ln energy transfer (see later). The Ln···Ln distances to that obtained in the single crystals.
Eur. J. Inorg. Chem. 2007, 291–301
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
293

P. C. Junk, M. M. Lezhnina et al.
FULL PAPER
The Tb3+ analogue of this series showed significant and only in refluxing DMF, albeit to a high cost in luminescence
exploitable luminescence in the visible, while Eu3+ is efficiency.
quenched by charge-transfer states and MLCT,[13] addition-
ally supported by vibrational quenching.[14] Gd3+ is known
to promote ligand phosphorescence at 24445 cm–1,[15] we
observe corresponding broad emission bands in Gd(paba)3
2.2. Co-coordinated Complexes
at 420 nm and 478 nm, however, accompanied by a compa-
rably strong emission due to minute Tb3+ impurities (see tives of the paba series above, we co-coordinated [Tb-
below). (paba)3H2O] with the bidentate ligands bipyridine and
To explore further the luminescence properties of deriva-
The luminescence of Tb(paba)3 relies on the following phenanthroline as well as the tridentate terpyridine. The
successive steps:[16,17] 1. singlet–singlet absorption of the li- complexes were accessed by crystallisation from solutions
gand (1S0 Ǟ1S*, π–π in nature), 2. intersystem crossing of pabaH and bipy, phen, or terpy, respectively, in ethanol,
(ISC, 1S*Ǟ3T, promoted by the spin-orbit coupling due to to which aqueous TbCl3 was added. In order to identify the
the Tb3+ ion), 3. (exchange resonance driven) energy trans- immediate Tb3+ coordination environment in the complexes
fer Ligand(3T)ǞTb3+(5D4). Emission, excitation end re- obtained {[Tb(paba)3(bipy)(H2O)]2·2H2O (5), [Tb(paba)3-
flectance spectra of solid Tb(paba)3H2O are given in Fig- (phen)(H2O)]2·2H2O (6), [Tb(paba)3(terpy)(H2O)]2·6.7H2O
ure 3.
(7)}, single crystals grown from these solutions by slow
As may be seen clearly, excitation can most efficiently be evaporation were subjected to single-crystal structure analy-
accomplished in this complex at a comparably short wave- ses. The X-ray crystal structure of [Tb(paba)3(bipy)-
length of approximately 310 nm. The quantum yield of the (H2O)]2·2H2O was previously reported.[18] For comparison,
complex amounts to an unexpectedly high value (88%, e.g. the La and Eu and analogues involving 1,10-phenanthroline
λexc = 340 nm), which is very surprising in view of the coor- were also structurally investigated. It is of importance to
dination of both, -NH2 and OH2 as ligands, as they are understand the structural variation of the rare earth series
usually both essential vibrational deactivators of the excited when investigating their luminescence properties. For exam-
state. However, it should be noted that both the Ln–OH2 ple, we have shown that in the coordination polymer
distance (261 pm) and the Ln–NH2 distance (276 pm) are Na[Tb(pic)4]·1.5H2O (picH = 2-picolinic acid) one-dimen-
fairly long and multiphoton emission may thus be of sec- sional energy migration occurs, which can be utilised to
ondary importance only, which is even more pronounced transfer energy to Eu3+ if the latter is co-doped into the
due to fairly strong hydrogen bonding.
material.[19] However, co-doping is only successful in these
A significant shortcoming for further utilisation of the compounds, if the structural integrity of the parent com-
complexes is that their strong degree of association [see pound is not compromised, a factor, which we have ob-
Figure 2, b)] prevents, for example, evaporation without served in the above-mentioned picolinates, if lanthanum
decomposition or dissolution in aprotic organic sol- was introduced. Therefore, we typically address the struc-
vents. We could obtain solutions from [Tb(paba)3H2O] tural characteristics of several of the lanthanoid series.
Figure 3. Comparison of optical spectra of the free complex Tb(paba)3H2O (powder, solid line) and the polymer composite Tb-
(paba)3PU (monolith, broken line).
294
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 291–301

Optically Functional Polyurethane Composites
FULL PAPER
Therefore, we also present the X-ray structural characterisa- terpy molecule, a monodentate water molecule, a bidentate
tion of [Eu(paba)3(phen)(H2O)]2·2H2O (8) and [La(paba)3- carboxylate group, a monodentate carboxylate group and a
(pabaH)(phen)2]·H2O (9). Additional support for the signi- monodentate carboxylate group that bridges to the other
ficance of detailed structural data will be provided at the Tb atom (Figure 5). The Tb···Tb distance in this dimer is
end of this paragraph.
5.312(8) Å.
In the published structure of [Tb(paba)3(bipy)(H2O)]2·
2H2O (5) the compound exists as a dimer, with each ter-
bium atom eight-coordinate. Each Tb atom is bound by a
bidentate bipy ligand, a monodentate water molecule and
three carboxylates, each with different binding modes. One
carboxylate is chelating bidentate, the second is a mono-
dentate-bound ligand, while the third bridges to the other
Tb atom. There is a water molecule in the lattice and the p-
amino groups of the paba ligands are not coordinated to
the Tb atoms. The Tb···Tb distance in the dimer is
5.554(3) Å, thus allowing energy migration between the
ions within the dimer.
[Tb(paba)3(phen)(H2O)]2·2H2O (6) crystallises in the tri-
¯
clinic space group P1 with one whole monomer comprising
the asymmetric unit. The compound dimerises through hy-
drogen bonding from a Tb-bound water molecule to the O
atom of a Tb-bound carboxylate group (Figure 4). Each Tb
atom is nine-coordinate, being bound by a bidentate 1,10-
phenanthroline ligand, three chelating bidentate carboxyl-
ate groups and a water molecule. There are also two lattice
water molecules per hydrogen-bonded dimer. The Tb···Tb
distance in the hydrogen-bonded dimer is 5.884(5) Å.
[Eu(paba)3(phen)(H2O)]2·2H2O (8) is isomorphous with 6
(Figure 4).
Figure 5. X-ray crystal structure of dimeric, hydrogen-bonded
[Tb(paba)3(terpy)(H2O)]2·H2O (7); hydrogen atoms on the aromatic
groups are omitted for clarity. Thermal ellipsoids are at the 50%
probability level.
[La(paba)3(pabaH)(phen)2]·H2O (9) crystallises in the
monoclinic space group P21/c with one whole mononuclear
unit and one lattice water molecule comprising the asym-
metric unit of the structure. The nine-coordinate lanthanum
centre is now bound by two bidentate 1,10-phenanthroline
ligands, one bidentate carboxylate, two unidentate carbox-
ylate and one unidentate carboxylic acid molecule (Fig-
ure 6). It is significant that the lanthanum centre is now
bound by a unidentate carboxylic acid molecule rather than
by water (or ethanol) molecules, from which the compound
was crystallised.
Figure 4. X-ray crystal structure of dimeric, hydrogen-bonded
[Tb(paba)3(phen)(H2O)]2·2H2O (6); hydrogen atoms on the aro-
matic groups are omitted for clarity. Thermal ellipsoids are at the
50% probability level. [Eu(paba)3(phen)(H2O)]2·2H2O (8) is iso-
structural.
Figure 6. X-ray crystal structure of mononuclear [La(paba)3-
(pabaH)(phen)2]2 (9); hydrogen atoms on the aromatic groups are
omitted for clarity. Thermal ellipsoids are at the 50% probability
level.
In compound 8 the Eu···Eu distance is 5.909(5) Å and is
slightly longer due to the larger ionic radius of Eu over
Tb.[10]
[Tb(paba)3(terpy)(H2O)]2·6.7H2O (7) crystallises in the
¯
triclinic space group P1 with one whole [Tb(paba)3(ter-
The proton on the bound carboxylic acid molecule is in-
py)(H2O)]2 dimer and another half dimer (as well as 10 lat- volved in hydrogen bonding with one of the O atoms of
tice water molecules) comprising the asymmetric unit. Each a unidentate bound carboxylate group. The fact that this
Tb atom is nine-coordinate, being bound by a tridentate compound is a new structural type for this chemistry, and
Eur. J. Inorg. Chem. 2007, 291–301
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
295

P. C. Junk, M. M. Lezhnina et al.
FULL PAPER
also that it is mononuclear is exciting, because we now in- high a quantum efficiency is not maintained in [Tb-
tend to expand this chemistry to the Eu and Tb analogues, (paba)3(phen)(H2O)]2·2H2O, where a drop to 38% is mea-
where luminescence may be enhanced due to decreased pos- sured at 380 nm.
sibility of Ln···Ln quenching. The shortest La···La distance
in compound 9 is 10.594(2) Å.
It was furthermore striking to us to note that the short
Ln···Ln distances of 4.725(3) Å found in [Gd(paba)3H2O]
In all compounds comprising the bidentate bipy or phen have an immediate impact on the luminescence spectra: ter-
or tridentate terpy, bond lengths are unexceptional between bium impurities (sub-percent level, Ͻ 0.01%) in the Gd
ligands and metal centres. A noteworthy feature of these complex lead to a strong Tb3+ emission, which had approxi-
compounds, when compared with the parent [Ln(paba)3- mately 10 times the intensity of the corresponding ligand
(H2O)] compounds, is that the para-nitrogen atoms of the phosphorescence. Characteristically, the sensitisation of
paba group are not involved in binding to the lanthanoid Tb3+ is already 4 times weaker in [Gd(paba)bipy(H2O)2]
centre (cf. above).
(Ln···Ln ca. 5.5 Å) and is not detectable in [Gd(paba)-
The optical properties of [Tb(paba)3(bipy)(H2O)]2·2H2O phen(H2O)2](H2O)1.5. (Ln···Ln ca. 5.9 Å). This behaviour
and [Tb(paba)3(phen)(H2O)]2·2H2O are depicted in Fig- signals the importance of detailed structural data in the in-
ure 7. In comparison to [Tb(paba)3·H2O]n the excitation terpretation of luminescent properties. Unfortunately, suit-
maximum has been red-shifted by approximately 50 nm on able crystals of the Gd analogue to [La(paba)3(pab-
bipy-co-coordination and about 70 nm on phen-co-coordi- aH)(phen)2](H2O) with a La···La distance 10.594(2) Å
nation [Tb(paba)3(bipy)(H2O)]2·2H2O.
could not be obtained as yet, but may in the future be an
This redshift is of high applicatory interest; because the interesting system to evaluate the details of Ln···Ln energy
excitation now falls in the range of commercially available transfer in rare earth metallo-organic complexes.
and ever improving semiconductor LED excitation sources,
and may thus be used in a broader range of low-cost de-
vices. Furthermore, the quantum yield of [Tb(paba)3(bi-
py)(H2O)]2·2H2O now even amounts to 95% under exci-
2.3. Reactions with Isocyanates
tation at 350 nm. This somewhat unusual result may, just
The amino group of the paba complexes appeared to be
as argued for [Tb(paba)3H2O] above, once more be interpre- a suitable reaction centre for the attachment of the com-
ted in terms of the molecular structure, displaying long Tb– plexes to polymeric backbones, isocyanates and polyure-
OH2 distances and hydrogen bonding. Unfortunately, that thanes (“PUs”) in particular. As suitable polymer matrices,
Figure 7. Comparison of free and polymer-imbedded complexes. Top, solid line: Tb(paba)3phen; broken line: Tb(paba)3phenPU. Bottom,
solid line: Tb(paba)3bipy; broken line: Tb(paba)3bipyPU. i) excitation, ii) emission ii) reflectance of pure complexes; iv) excitation, v)
emission, vi) reflectance of PU composites.
296
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 291–301

Optically Functional Polyurethane Composites
FULL PAPER
hexamethylene diisocyanates [O=C=N–(CH2)6–N=C=O, ethane matrix, based on HDI and a polyol, in which doping
“HDIs”] and derived prepolymers were chosen rather than with Tb(paba)3 substitutes a corresponding amount of the
e.g. methylenebis(1,4-phenylene) diisocyanate (O=C=N– polyol [Equation (9)]. Analogously, Tb(paba)3bipy and
C6H4–CH2–C6H4–N=C=O, “DPDIs”) or 1,4-phenylene di- Tb(paba)3phen have been reacted with HDI and polyol to
isocyanate (O=C=N–C6H4–N=C=O, “PDIs”), because give Tb(paba)3bipyPU and Tb(paba)3phenPU. The “con-
their optical transparency still allows optical measurements centrated” polymeric Tb(paba)3PDI, Tb(paba)3DPDI and
of the complexes imbedded, which might otherwise be ob- Tb(paba)3HDI are obtained as white powders, which are
scured due to the extended π electrons, if DPDIs or PDIs insoluble in all polar and apolar solvents tested; it was not
were chosen as the matrix. However, reactions of Tb(paba)3 possible to obtain transparent layers or monoliths from
with pure HDI, PDI and MDI were tested as “concentrated these materials. The excitation spectra of these reaction
polymers” without further crosslinkers [see reaction products are represented in Figure 9 in comparison to
schemes Figure 8, Equation (8)]. In the following, Tb(paba)3; the corresponding emission spectra only differ
“Tb(paba)3HDI”, “Tb(paba)3PDI” and “Tb(paba)3DPDI” in intensities and are not reproduced. Evidently, PDI link-
exclusively designate the reaction products of Tb(paba)3 age leads to a slight increase in intensity accompanied by a
with HDI, PDI and DPDI, respectively, according to Equa- redshift of the excitation maximum to approximately
tion (8), while “Tb(paba)3PU” refers to a complete polyur- 330 nm. Unfortunately, due to cloudiness (i.e. light scat-
Figure 8. Reaction schemes of Tb(paba)3 and diisocyanates formation of “concentrated” polymeric Tb(paba)3 diisocyanates, Equation
(8). Formation of PU-polymer “doped” with Tb(paba)3, Equation (9).
Figure 9. Reaction products of Tb(paba)3 with different diisocyanates. Left: excitation spectra. Right: Photodegradation. i) Tb-
(paba)3PDI; ii) Tb(paba)3; iii) Tb(paba)3DPDI; iv) Tb(paba)3HDI, see text.
Eur. J. Inorg. Chem. 2007, 291–301
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
297

P. C. Junk, M. M. Lezhnina et al.
FULL PAPER
tering), shape and morphology of the polymer layers, reli- quantum yields of Tb(paba)3phenPU, but from the exci-
able data on quantum efficiencies could not be determined, tation spectra, they may be estimated to amount to Ͼ 40%
however, a crude visual comparison with diluted samples of under 260 nm excitation.
Tb(paba)3 in PU [Tb(paba)3PU, see following paragraph]
suggests comparable efficiencies.
For eventual applications of the polymers described,
knowledge on further material parameters next to the op-
Transparent composite materials, Tb(paba)3PU, could tical performance will be required, among them the durabil-
eventually be prepared using DMF solutions of Tb(paba)3 ity and stability of the emission intensity. We have therefore
in a polyol (see experimental section), to which HDI was also invested some effort into the characterisation of the
added [Equation (9)] in Figure 8, R = (CH2)6. As opposed emissive properties under prolonged irradiation with UV
to that, for reasons not obvious to us, the reactions of po- light. For this purpose, pressed sample polymer disks of
lyol-dissolved Tb(paba)3 with PDI and DPDI only gave tur- approximately uniform thickness of approximately 1 mm to
bid products. The optical spectra of transparent Tb(paba)3- grant complete absorption were irradiated and the deprecia-
PU are given in Figure 3. The comparison with pure tion of the Tb3+ 545 nm emission line (5D4 Ǟ7F5) was mon-
Tb(paba)3 immediately demonstrates the shortcomings of itored. For comparison, pure Tb(paba)3 was pressed to a
the composite. The onset of absorption is red-shifted to ap- powder disk and included in the series. The results of the
proximately 400 nm due to matrix contributions, which do ageing experiments on Tb(paba)3PDI, Tb(paba)3DPDI and
not produce a noteworthy emissive response down to Tb(paba)3HDI and Tb(paba)3 are contained in Figure 9.
320 nm, while the excitation maximum is blueshifted to
The evaluation of the depreciation of Tb(paba)3PDI,
260 nm. The overall emission quantum yield is thus reduced which shows the highest initial efficiency, reveals that a
by a factor of two to four (42%, λexc = 260 nm; 17%, λexc maintenance crossover with pure Tb(paba)3 is reached after
= 300). However, some error in this value may arise from only a few minutes, indicating that the aromatic polymer
differing sample and reference shapes (see Exp. Sect.). Cor- backbone of PDI is liable to relatively fast photodegrad-
respondingly, the extended conjugation brought about by ation. Similarly, Tb(paba)3DPDI shows very similar degra-
the –C6H4–NH–CO–NH– substituents [see Figure 8, Equa- dation behaviour, albeit at a lower efficiency level. Remark-
tion (2)] may be held responsible for the redshift of the ab- ably, the aliphatic HDI based polymer is comparably stable,
sorption. At the same time, the coordination numbers of even more so than the pure complex. All curves could be
six for the Tb3+ ion, as sketched for simplicity in Figure 8, fitted as single exponentials, if the initial degradation delay
is most unlikely in the polymer. Instead, coordination of N observed for Tb(paba)3 and Tb(paba)3HDI was taken into
and O atoms of neighbouring fragments may be expected, account. Based on extrapolation, the maintenance level of
thus giving rise to radiationless deactivation by high fre- Tb(paba)3HDI can accordingly be estimated to amount to
quency vibrations.
almost 80% after 10000 h of operation. Future efforts will
The initial idea of using bipyridine and phenanthroline thus be directed towards luminescence improvement of ali-
as co-ligands to increase their solubility in the (pre-)poly- phatic polyurethanes.
mers did not lead to the desired results with respect to effi-
ciency. Although bipyridine co-coordinated Tb(paba)3
could eventually be used to form
[“Tb(paba)3bipyPU”], if the Tb(paba)3bipy was initially
a clear polymer
Conclusions
dissolved in excess HDI followed by addition/crosslinking
Stimulated by high luminescence efficiencies of p-amino-
with polyol, its efficiency is again lowered by a factor of benzoates of Tb3+ (quantum yields Ͼ 80%), detailed struc-
two to three (55%, λexc = 260 nm to 28%, λexc = 300 nm) tural investigations have been performed in order to derive
in comparison to pure Tb(paba)3. If Tb(paba)3bipy was dis- materials from Tb(paba)3, which are applicable in photolu-
solved in the polyol prior to HDI addition, bipyridine ap- minescenct applications, e.g. luminescent markers, or,
peared to be liberated, as indicated by the excitation spec- eventually electrooptical devices, such as OLEDs. Typically,
trum, which is then again in resemblance of Tb(paba)3PU, in the series of rare earth p-aminobenzoates coordination
albeit at much lower intensity, at least partly due to the polymers are formed, which were broken down to molecu-
radiation “lost” to bipy absorption. This behaviour is yet lar species for further manufacture. Such can in part be ac-
more pronounced and evident in Tb(paba)3phen, where a complished using chelating ligands such as bipyridine,
blue emission of phen indicates the presence of the uncoor- phenanthroline, or terpyridine, although crystal structure
dinated co-ligand. The emission of phen is here seen to a determinations reveal that they essentially still exist as di-
similar extent, regardless of the addition sequence (dissol- meric units and disadvantageously contain coordinated
ution in polyol followed by addition of HDI, or dissolution water. Nevertheless, on co-coordination of with bipyridine,
in HDI, followed by polyol admixture). The loss of phen is a slight increase in quantum yield is accomplished (Ͼ 90%)
clearly visible from the spectral agreement of Tb(paba)3PU in the case of Tb3+, while phenanthroline reduces the over-
(broken line in Figure 3) with Tb(paba)3phenPU (broken all efficiency (36%). The p-aminobenzoates of Tb3+ were
line in Figure 7, a); the substitution of the bidentate co- linked to aliphatic and aromatic diisocyanates through the
ligands by the polydentate polymer fragments may then be benzoate amino group, yielding brightly green luminescing,
interpreted in terms of an entropy increase. Due to the blue insoluble polymer composites. Consecutively, transparent
ligand emission, we refrained from the determination of polyurethane monoliths incorporating backbone-linked p-
298
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 291–301

Optically Functional Polyurethane Composites
FULL PAPER
aminobenzoates of Tb3+ and co-ligated complexes could be
fabricated using aliphatic diisocyanates, and polyols as ad-
ditional components. These “doped” polyurethanes show
photoluminescence with quantum yields in the range be-
tween 40 and 55%, albeit at lowered excitation wavelengths.
In anticipation of potential applications, the photodegrad-
ation of some aromatic and aliphatic composites were
evaluated. In these experiments, aliphatic polymer back-
bones proved to be superior, but had lower initial efficienc-
ies. Future work will thus focus on increasing the p-amino-
benzoate composite efficiencies. Additionally, we are cur-
rently testing other benzoic acid derivatives.
7.72 (s), 10.00 (s), 14.88 (m), 15.56 (m), 18.38 (w), 18.76 (w), 19.46
(w), 20.04 (m), 21.62 (w), 22.34 (m), 23.74 (m), 24.12 (m), 24.92
(m), 27.44 (w), 28.54 (w), 28.98 (w), 29.52 (w), 30.04 (w), 30.42 (w),
31.06 (w), 32.80 (w), 33.74 (w), 35.48 (w), 36.62 (m), 37.22 (w),
38.56 (w), 39.80 (w), 40.58 (w), 42.10 (m), 44.00 (w), 44.72 (m),
45.70 (w), 47.82 (w).
[Tb(paba)3(H2O)2]2·2H2O (4): From one attempted synthesis of co-
crystallising [Tb(paba)3(H2O)] with phenanthroline, we obtained
crystals of the title compound that were isostructural with the pre-
viously reported complex.[9d] Unit cell data for compound 4: tri-
¯
clinic, space group P1 (No. 2), a = 10.7819(7), b = 12.0981(8), c =
13.2695(9) Å, α = 72.407(1), β = 66.679(1), γ = 87.981(1)°, V =
1508.0(2) Å3.
[Tb(paba)3(bipy)(H2O)]2·2H2O (5) and [Tb(paba)3(phen)(H2O)]2·
2H2O (6), [Tb(paba)3(terpy)(H2O)]2·6.7H2O (7), [Eu(paba)3(phen)-
(H2O)]2·2H2O (8) and [La(paba)3(pabaH)(phen)2].H2O (9),
[Gd(paba)3(bipy)(H2O)]2·2H2O (10) [Gd(paba)3(phen)(H2O)]2·
2H2O (11): [Tb(paba)3(bipy)(H2O)]2·2H2O was prepared by adding
an aqueous solution of TbCl3 (0.25 g) to an ethanol/water
(C2H5OH/H2O, 1:1) solution of p-aminobenzoic acid (0.2 g) and
bipyridine (0.3 g), to give a volume of ca. 30 mL. The colourless
crystals were obtained by slowly evaporating the mixture at room
temperature for several days.
Experimental Section
Lanthanoid salts were purchased from Rhodia, Chempur and Ald-
rich with specified purity levels of in excess of 99.9%; p-aminoben-
zoic acid was obtained from Aldrich. All starting materials were
used without further purification. Deionised water was used in all
syntheses. Infrared spectra were recorded on sodium chloride plates
as Nujol mulls or pressed into KBr disks (Figure 1) on a Perkin–
Elmer Spectrum One FTIR spectrometer. Lanthanoid metal analy-
ses were determined titrimetrically using the Na2EDTA method.
Carbon analyses were carried out using a C-analyser with IR-detec-
tion unit (ELTRA, Germany). Melting points were performed in
unsealed tubes.
Analogously, the preparation of [Ln(paba)3(phen)(H2O)]2·2H2O,
[Ln = Eu, (8) Tb (6), Gd (11)] involved the addition of an aqueous
LnCl3 (Eu, Tb, Gd) solution (0.2 g) to an ethanol solution of p-
aminobenzoic acid (0.3 g) and 1,10 phenanthroline (0.3 g) after two
weeks in an open vessel afforded small light yellow crystals suitable
for single crystal structure determination.
Synthesis: The complexes 1–3 were synthesised following a similar
procedure. Thus, sodium p-aminobenzoate (obtained by the reac-
tion of the free p-aminocarboxylic acid with sodium hydrogencar-
bonate) was added to and aqueous solution of lanthanoid chloride.
The crystals were obtained from slightly acidic aqueous medium.
The samples were evaporated until large crystals of the complexes
deposited. Complex 4 was synthesised in a similar manner, except
1,10-phenanthroline was added to the reaction mixture. Details of
the properties for each compound follow. Yields reported are for
isolated crystalline material. Yields of these reactions can be en-
hanced by precipitation of the compounds by addition of acetone
to a saturated aqueous solution or increasing the pH value to just
below 8 with 2  ammonia.
[Eu(paba)3(H2O)] (1): VEu: 2.43 cm3 (0.52 mol/dm3) = 1.26 mmol.
mH–p–ABA: 0.52 g = 3.79 mmol. mproduct: 0.68 g (yield 93.0%).
C21H20EuN3O7: calcd. C 43.6, Eu 26.3; found C 43.4; Eu 26.5.
FTIR: see Figure 1. Powder XRD: 7.72 (s), 9.96 (s), 15.56 (m),
18.50 (w), 20.04 (m), 22.38 (w), 23.74 (m), 24.16 (m), 24.98 (w),
27.50 (w), 28.54 (w), 29.48 (w), 30.02 (w), 30.96 (w), 32.52 (w),
33.62 (w), 36.78 (m), 37.08 (m), 38.46 (s), 39.68 (w), 40.64 (w),
42.00 (w), 42.92 (w), 44.04 (w), 44.72 (s), 45.76 (w), 46.70 (w), 47.60
(w), 49.42 (w). DTA: 200 °C (exothermic), 390 °C (endothermic),
430 °C (endothermic), 500 °C (endothermic), 740 °C (endother-
mic). DTG: 120–220 °C (3% weight loss), 330–460 °C (30%), 460–
810 °C (69% weight loss).
“Working amounts” of 5, 6, 10 and 11 were obtained using
H(paba) (3 mmol) and bipy (or phen) (2 mmol), dissolved in of
water/ethanol mixture (20 mL, 1: 1), adjustment of the pH to 5
using 2  ammonia solution. An aqueous solution of Tb3+ [1 mmol
of from of 0.5  TbCl3 (GdCl3) stock solution] was then slowly
added to the ligand solution under vigorous stirring. Almost com-
plete and immediate precipitation occurs after slowly raising the
pH of this solution to just below 8.0. The complexes were filtered
and dried at 60 °C. The products deviated in their composition
from the material used for crystal structure determination due to
the drying procedure (loss of lattice water): [Gd(paba)3bipy-
(H2O)2]: calcd. C 48.95; found 48.96; [Gd(paba)3phen(H2O)2]-
(H2O)1.5: calcd. C 48.83; found C 48.90].
For the synthesis of crystals of [Tb(paba)3(terpy)(H2O)]2·6.7H2O
(7) terpyridine (0.3 g) and p-aminobenzoic acid (0.2 g) were dis-
solved in 50% ethanol (3 mL). To the resulting clear solution,
TbCl3 solution (0.2 g in 3 mL water) was added dropwise. The reac-
tion was stirred for 5 h. A few light-yellow crystals were received
on partial evaporation of the solvent.
Single crystals of [La(paba)3(pabaH)(phen)2]·H2O (9) were pre-
pared as for compounds 6 and 8 except LaCl3 solution was used
in place of LnCl3(Ln = Eu, Tb).
[Gd(paba) (H O)] (2): Yield 17%. IR (Nujol): ν = 3400 m (br),
˜
3
2
Tb(paba)3HDI, Tb(paba)3PDI and Tb(paba)3DPDI: The complex
polymer composites were synthesised by adjusting a procedure de-
scribed in the literature.[20] All operations were carried out provid-
ing for the exclusion of water (standard Schlenk techniques). Vac-
uum dried Tb(paba)3 (10–3 mbar, 140 °C, 5 h, 0.010 mol) was dis-
solved in dry DMF (20 mL). To this solution the pure diisocyanate
(0.017 mol) was added dropwise. The reaction was carried out
within 14 h at 80 °C under argon. After solvent evaporation, the
residue was dried in vacuo and powdered using a mortar and pestle.
1660 s, 1595 s, 1567 s, 1399 m, 1347 s, 1306 w, 1297 m, 1260 w,
1240 w, 1163 m, 1149 w, 1093 w, 1051 w, 1013 w, 848 w, 787 m,
760 m, 698 s, 634 m cm–1. C21H20GdN3O7: calcd. C 43.00; found
C 43.08.
[Tb(paba)3(H2O)] (3): VTb: 2.30 cm3 (0.58 mol/dm3) = 1.34 mmol.
mH–p–ABA: 0.55 g = 4.01 mmol
mproduct: 0.72 g (yield 92.0%). C21H20N3O7Tb: calcd. C 43.1, Tb
27.3; found C 43.9, Tb 27.2.. FTIR: see Figure 1. Powder XRD:
Eur. J. Inorg. Chem. 2007, 291–301
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
299

P. C. Junk, M. M. Lezhnina et al.
FULL PAPER
Tb(paba)3PU, Tb(paba)3bipyPU and Tb(paba)3phenPU: For the
preparation of the transparent polyurethanes a commercial system
consisting of a hexamethylene diisocyanate prepolymer and a po-
lyol was used (from Lackwerke Peters, 47906 Kempen, Germany,
VT 3402 KK). To determine the exact amounts of polyol to be
substituted by the corresponding Tb complexes, the NCO content
was determined using chlorobenzene as solvent, dibutylamine as
reagent and bromothymol blue as the indicator for the HDI, for
the OH number of the polyol, pyridine, phthalic anhydride and
phenolphthalein were used, according to procedures previously de-
scribed.[21] Analytically, 5.40 mmol NCO/g diisocyanate and
5.34 mmol OH/g polyol were found.
filter (from Schott, Germany, UG11, 3 mm). The change of emis-
sion was monitored through a green interference filter (from LOT-
Oriel, 64293 Darmstadt, Germany, 546FS10-50) and focussed on a
CCD camera (from Acton Research, Spectrum MM) through a
Nikon® lens at a distance of 25 cm. Light source, filters and detec-
tion unit were attached to a housing, which contained the samples
at the bottom. A freshly prepared Al mirror (thermal evaporation,
25 cm2) was used to turn the horizontally entering beam onto the
samples at the bottom, the camera was aligned on top of the hous-
ing. Background radiation was accounted for by subtraction of
BaSO4 measurements prior to interpretation.
Xray Crystallography Studies: For all compounds X-ray quality
crystals were sealed and mounted in thin-walled capillaries, with
hemispheres of data collected at room temperature with a Bruker
To obtain dilute samples for efficiency measurements, the com-
plexes (2 mmol) were dissolved in DMF and added to the hexa-
methylene diisocyanate (1 g, ratio paba: NCO group approximately SMART CCD diffractometer (Mo-Kα radiation, λ = 0.71073 Å)
1:100), DMF was evaporated in vacuo and the complex solution
in diisocyanate thus obtained was mixed with the polyol (1:1 ratio),
the mixture was placed in a transparent cuvette of 1 cm path
length, air bubbles were removed under vacuum. The polymerisa-
tion to give the polyurethane was carried out at 60 °C in a drying
chamber. Polymers containing more than 1·10–2 mol complex/L
could be prepared in an analogous manner. For such samples, the
consumption of HDI by Tb(paba)3 or was accounted for in that a
lower amount of polyol was used. However, the emission intensities
(not quantum yields) were at a maximum at approximately
1·10–2 mol complex/L.
using the omega scan mode. Data sets were corrected for absorp-
tion using the program SADABS.[22] For all structures, the position
of the heavy atoms were found using the Patterson method for
heavy atoms and refined on F2 using SHELXL97–2[23] with X-
SEED as the graphic interface.[24] All non-hydrogen atoms were
located and were refined with anisotropic thermal parameters. Hy-
drogen atoms were placed in calculated positions (riding model)
and were not refined. For compounds 2, 6, 8 and 9 hydrogen atoms
were located on the NH2 groups and on water molecules and were
refined isotropically. For compound 7 hydrogen atoms were not
located on heteroatoms and were not included in the refinement.
Crystal data, and a summary of data collection are listed below.
Optical Characterisation: All emission and excitation intensities in
the spectra reproduced are scaled against each other, such as to
grant direct comparability.
CCDC-607965 (for 2), -607966 (for 6), -607967 (for 7), -607968 (for
8) and -607969 (for 9) contain the supplementary crystallographic
data for this paper. These data can be obtained free of charge
from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Emission and excitation spectra of the solids were recorded at an
angle of incidence of ca. 50° using a praying mantis type of set up
to allow horizontal sample positioning without a cover. Reflectance
spectra were obtained using an integrating sphere with a diameter
of 70 mm and 10 mm port openings vs. commercial white and black
references (Labsphere), and BaSO4, respectively. Two Acton 300
Monochromators of 300 mm focal length were thus synchronously
scanned. The intensity of the 450 W Xe source was detected with
an Acton Photomultiplier tube P2, the gratings had 1200 gmm–1.
The excitation spectra of Tb(paba)3HDI, Tb(paba)3PDI and
Tb(paba)3DPDI presented in Figure 9 were corrected for the lamp
and spectrometer profiles with BaMgAl10O17: Eu (“BAM”).
Tb(paba)3, Tb(paba)3bipy, Tb(paba)3phen were also measured
against Lumogen Red doped PMMA powders (doping level
50 ppm), whose quantum yield was taken as a constant 42% in the
excitation range between 250 nm and 500 nm. Relative quantum
yields of the clear PU composite monoliths [Tb(paba)3PU,
Tb(paba)3bipyPU, Tb(paba)3phenPU], prepared in 1 cm cuvettes,
were determined relative to a Lumen Red/PMMA disk of 3 mm
thickness. Due to the confinement in the reference disk of lower
thickness, some loss due to lateral propagation of emitted light
within the disk is possible, but could not be accounted for, because
neither sample cuvette nor reference disk could be inserted into the
integrating sphere. Additionally, we would like to point out that
the quantum efficiencies given, while reproducible within our op-
tical setup, may still carry imprecisions, not the least owing to the
lack of a suitable green emitting standard with continuous quan-
tum yield in the spectral excitation range between 250 and 500 nm
(estimated imprecision margin Ϯ10%). However, the general trends
described and discussed will not be affected.
Crystal Data for Compound 2: C21H20GdN3O7, M = 583.65,
0.24ϫ0.18ϫ0.15 mm, monoclinic, space group P21/n (No. 14), a
= 9.7452(9), b = 22.736(2), c = 9.8229(9) Å, β = 99.9290(10)°, V =
2143.9(3) Å3, Z = 4, Dc = 1.808 g/cm3, F(000) = 1148, 9744 reflec-
tions collected, 3072 unique (Rint = 0.0665). Final GooF = 1.045,
R1 = 0.0287, wR2 = 0.0691, R indices based on 2639 reflections
with IϾ2σ(I) (refinement on F2), 323 parameters, 0 restraints. Lp
and absorption corrections applied, µ = 3.142 mm–1.
Crystal Data for Compound 6: C33H30N5O8Tb, M = 783.54,
¯
0.36ϫ0.35ϫ0.32 mm, triclinic, space group P1 (No. 2), a =
10.4554(8), b = 12.3129(9), c = 14.0035(10) Å, α = 92.8650(10), β
= 102.8890(10), γ = 109.0390(10)°, V = 1646.4(2) Å3, Z = 2, Dc =
1.581 g/cm3, F(000) = 784, 7545 reflections collected, 4665 unique
(Rint = 0.0461). Final GooF = 1.050, R1 = 0.0466, wR2 = 0.1283,
R indices based on 4260 reflections with IϾ2σ(I) (refinement on
F2), 466 parameters, 15 restraints. Lp and absorption corrections
applied, µ = 2.206 mm–1.
Crystal Data for Compound 7: C230H228N38O66Tb6, M = 5534.02,
¯
0.32ϫ0.30ϫ0.15 mm, triclinic, space group P1 (No. 2), a =
17.460(3), b = 18.615(3), c = 19.377(3) Å, α = 80.199(3), β =
72.342(3), γ = 73.823(3)°, V = 5738.3(14) Å3, Z = 1, Dc = 1.601 g/
cm3, F(000) = 2792, 26873 reflections collected, 16542 unique (Rint
= 0.0803). Final GooF = 1.011, R1 = 0.0768, wR2 = 0.1882, R
indices based on 9943 reflections with IϾ2σ(I) (refinement on F2),
1531 parameters, 0 restraints. Lp and absorption corrections ap-
plied, µ = 1.918 mm–1.
The photochemical degradation data of the samples presented in
Figure 9 were collected under irradiation with a 150-W Xe lamp
that was shone onto an area of 400 cm2 through a UV band pass
Crystal Data for Compound 8: C33H30EuN5O8, M = 776.58,
0.20ϫ0.20ϫ0.20 mm, triclinic, space group P1 (No. 2), a =
10.4543(11), b = 12.2021(13), c = 14.0607(15) Å, α = 92.904(2), β
¯
300
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 291–301

Optically Functional Polyurethane Composites
FULL PAPER
= 102.593(2), γ = 108.734(2)°, V = 1643.6(3) Å3, Z = 2, Dc
=
[6] A. Ouchi, Y. Suzuki, Y. Ohki, Y. Koizumi, Coord. Chem. Rev.
1998, 92, 29.
[7] B. Yan, H. Zhang, Mater. Res. Bull. 1998, 10, 1517.
[8] D. Sendor, P. C. Junk, U. Kynast, Solid State Phenomena 2003,
90/91, 521.
[9] a) M. S. Khiyalov, I. R. Amiraslanov, Kh. S. Mamedov, E. M.
Movsumov, Zh. Strukt. Khim. (J. Struct. Chem.) 1981, 22, 113;
b) S.-Y. Niu, Y.-L. Zhang, W. Lai, Z.-Z. Yang, G.-D. Yang, L.
Ye, Huaxue Xuebao 2001, 59, 2170; c) S. Y. Niu, G. D. Yang,
Y. L. Zhang, J. Jin, L. Ye, Z. Z. Yang, J. Mol. Struct. 2002,
608, 95; d) H.-L. Sun, C.-H. Ye, X.-Y. Wang, J.-R. Li, S. Gao,
K.-B. Yu, J. Mol. Struct. 2004, 702, 77; e) C.-H. Ye, H.-L. Sun,
X.-Y. Wang, J.-R. Li, D.-B. Nie, W.-F. Fu, S. Gao, J. Solid State
Chem. 2004, 177, 3735.
1.569 g/cm3, F(000) = 780, 7553 reflections collected, 4692 unique
(Rint = 0.0394). Final GooF = 1.071, R1 = 0.0537, wR2 = 0.1398,
R indices based on 4315 reflections with IϾ2σ(I) (refinement on
F2), 465 parameters, 15 restraints. Lp and absorption corrections
applied, µ = 1.966 mm–1.
Crystal Data for Compound 9: C52H43LaN8O9, M = 1062.85,
0.40ϫ0.40ϫ0.20 mm, monoclinic, space group P21/c (No. 14), a
= 18.8602(12), b = 19.7204(13), c = 13.1888(9) Å, β = 97.1610(10)°,
V = 4867.1(6) Å3, Z = 4, Dc = 1.450 g/cm3, F(000) = 2160, 22150
reflections collected, 6988 unique (Rint = 0.0646). Final GooF =
1.003, R1 = 0.0340, wR2 = 0.0648, R indices based on 5324 reflec-
tions with IϾ2σ(I) (refinement on F2), 676 parameters, 0 restraints.
Lp and absorption corrections applied, µ = 0.943 mm–1.
[10] R. D. Shannon, Acta Crystallogr., Sect. A 1976, 32, 751.
[11] Z. Rzaczynska, V. K. Belskii, Pol. J. Chem. 1994, 68, 309.
[12] L. Z. Rzaczynska, V. K. Belskii, Pol. J. Chem. 1994, 68, 309.
[13]
[14]
S. T. Frey, W. D. Horrocks Jr., Inorg. Chem. 1991, 30, 1073.
G. Blasse, B. C. Grabmeier, Luminescent Materials, Springer,
1994.
Acknowledgments
We acknowledge the Australian Research Council for continued
support. We also acknowledge an IPRS scholarship and Monash
Graduate Scholarship (to M.H.) and funding by the Bundesminis-
terium für Bildung und Forschung (BMB+F Grant 17 094 00) to
M.W. BAM and Lumogen Red references were generously provided
by Philips Research Laboratories (Aachen, Germany) and BASF
(Ludwigshafen, Germany), respectively.
[15]
B. Yan, S. B. Wang, J. Z. Ni, Monatsh. Chem. 1998, 129, 151.
[16] V. L. Ermolaev, V. G. Aleshin, E. A. Saenko, Dokl. Akad. Nauk
1965, 165, 1048.
[17] M. W. S. Sato, Bull. Chem. Soc. Jpn. 1970, 43, 1955.
[18] L. Oyang, H.-L. Sun, X.-Y. Wang, J.-R. Li, D.-B. Nie, W.-F.
Fu, S. Gao, K.-B. Yu, J. Mol. Struct. 2005, 740, 175.
[19] D. Sendor, M. Hilder, T. Juestel, P. C. Junk, U. H. Kynast, New
J. Chem. 2003, 27, 1070.
[20] C.-L. Lin, H.-M. Mao, R.-R. Wu, P.-L. Kuo, Macromolecules
2002, 35, 3083.
[21] Z. Wirpsza, Poliuretany, Chemia technologia zastosowanie,
WNT Warszawa, 1991.
[22] R. H. Blessing, Acta Crystallogr., Sect. A 1995, 51, 33.
[23] G. M. Sheldrick, University of Göttingen, 1997.
[24] L. J. Barbour, J. Supramol. Chem. 2001, 1, 189.
Received: August 29, 2006
[1] S. Sato, S. M. Wada, Bull. Chem. Soc. Jpn. 1970, 43, 1955.
[2] W. R. Dawson, J. L. Kropp, M. Windsor, J. Chem. Phys. 1966,
45, 2410.
[3] O. L. Malta, J. Lumin. 1997, 71, 229.
[4] F. R. G. e Silva, O. L. Malta, J. Alloys Compd. 1997, 250, 427.
[5] G. F. G. F. de Sá, O. L. Malta, L. de Mello Donegá, A. R.
Simas, R. L. Longo, P. A. da Silva Jr., E. F. Santa Cruz, Coord.
Chem. Rev. 2000, 196, 165.
Published Online: November 13, 2006
Eur. J. Inorg. Chem. 2007, 291–301
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
301